Etiology and epidemiology of diarrhea in children in Hanoi, Vietnam

International Journal of Infectious Diseases (2006) 10, 298—308

/>
Etiology and epidemiology of diarrhea in
children in Hanoi, Vietnam

Trung Vu Nguyen a,c, Phung Le Van a, Chinh Le Huy a,
Khanh Nguyen Gia b, Andrej Weintraub c,*

a Department of Medical Microbiology, Hanoi Medical University, Hanoi, Vietnam
b Department of Paediatrics, Hanoi Medical University, Hanoi, Vietnam
c Department of Laboratory Medicine, Division of Clinical Bacteriology, F-82, Karolinska Institutet,
Karolinska University Hospital, Huddinge, S-141 86 Stockholm, Sweden

Received 3 February 2005; received in revised form 17 May 2005; accepted 31 May 2005
Corresponding Editor: Richard Oberhelman, New Orleans, USA

KEYWORDS Summary
Diarrhea; Objectives: This paper provides a preliminary picture of diarrhea with regards to etiology,
Children; clinical symptoms, and some related epidemiologic factors in children less than five years of
Hanoi; age living in Hanoi, Vietnam.
Vietnam Methods: The study population included 587 children with diarrhea and 249 age-matched healthy
controls. The identification of pathogens was carried out by the conventional methods in
combination with ELISA, immunoseparation, and PCR. The antibiotic susceptibility was deter-
mined by MIC following the NCCLS recommendations.
Results: Of those with diarrhea, 40.9% were less than one year old and 71.0% were less than two
years old. A potential pathogen was identified in 67.3% of children with diarrhea. They were group
A rotavirus, diarrheagenic Escherichia coli, Shigella spp, and enterotoxigenic Bacteroides
fragilis, with prevalences of 46.7%, 22.5%, 4.7%, and 7.3%, respectively. No Salmonella spp or
Vibrio cholerae were isolated. Rotavirus and diarrheagenic E. coli were predominant in children
less than two years of age, while Shigella spp, and enterotoxigenic B. fragilis were mostly seen in

the older children. Diarrheagenic E. coli and Shigella spp showed high prevalence of resistance to
ampicillin, chloramphenicol, and to trimethoprim/sulfamethoxazole. Children attending the
hospitals had fever (43.6%), vomiting (53.8%), and dehydration (82.6%). Watery stool was
predominant with a prevalence of 66.4%, followed by mucous stool (21.0%). The mean episodes
of stools per day was seven, ranging from two to 23 episodes. Before attending hospitals, 162/587
(27.6%) children had been given antibiotics. Overall, more children got diarrhea in (i) poor
families; (ii) families where piped water and a latrine were lacking; (iii) families where mothers
washed their hands less often before feeding the children; (iv) families where mothers had a low
level of education; (v) families where information on health and sanitation less often reached
their households.

* Corresponding author. Tel.: +46 8 585 87831; fax: +46 8 711 3918.
E-mail address: (A. Weintraub).

1201-9712/$32.00 # 2005 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijid.2005.05.009

Diarrhea in children in Hanoi, Vietnam 299

Conclusions: Group A rotavirus, diarrheagenic Escherichia coli, Shigella spp, and enterotoxigenic
Bacteroides fragilis play an important role in causing diarrhea in children in Hanoi, Vietnam.
Epidemiological factors such as lack of fresh water supply, unhygienic septic tank, low family
income, lack of health information, and low educational level of parents could contribute to the
morbidity of diarrhea in children.
# 2005 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights
reserved.

Introduction Vietnam. They had not had any diarrheal episode for at least
one month before the collection of fecal samples.
Infectious diarrhea is a leading cause of morbidity and mor-

tality worldwide, affecting mainly infants.1 Approximately The children were enrolled in the study during a one-
12 million children in developing countries die before the age year period starting in March 2001 and ending in April 2002.
of five years, and 70% of these deaths are due to five health Diarrhea was characterized by the occurrence of three or
problems, including diarrhea.2 Unhygienic and unsafe envir- more loose, liquid, or watery stools or at least one bloody
onments place children at risk of death.3,4 Ingestion of loose stool within a 24-h period. An episode was considered
contaminated water, inadequate availability of water for resolved on the last day of diarrhea followed by at least
hygiene, and lack of access to sanitation contribute to about three diarrhea-free days. An episode was considered per-
1.5 million child deaths and around 88% of deaths from sistent if it continued for 14 or more days.16 Vomiting was
diarrhea per year.3,4 In addition, there are international defined as the forceful expulsion of gastric contents occur-
studies where it has been reported that a higher prevalence ring at least once in a 24-h period. Fever was defined as an
of diarrhea, with higher episodes of child diarrhea, are under-arm temperature of >37.2 8C. Thresholds of 37.2—
related to a low socio-economic status of the household 39 8C and >39 8C were set for moderate and high fever,
and community, as well as to a low educational level of respectively. Dehydration level was assessed following the
the child’s parents.5—8 recommendations of the WHO Program for Control of Diar-
rheal Diseases and these assessments were carried out by
Diarrhea is mainly caused by enteric pathogens including the pediatricians.17
viruses, bacteria, and parasites. Rotavirus and diarrhea-
genic Escherichia coli (DEC) are considered to be the most After informed consent was obtained, a pediatrician spe-
common of the many recognized enteropathogenic organ- cifically assigned to the study examined each patient and
isms, the former on a global scale,9 with DEC being parti- filled out the demographic data and information on clinical
cularly important in developing countries.10 Rotavirus, symptoms and illness onset on a standardized questionnaire.
especially group A rotavirus, is the leading cause of infantile The healthcare workers also obtained similar information
gastroenteritis worldwide and is responsible for approxi- from the controls.
mately 20% of diarrhea-associated deaths in children under
five years of age.11 Some other factors related to the demography and socio-
economic status of the children’s parents were also obtained.
There are six main categories of DEC identified. These are: Education of the parents was assessed as being at either a
(i) enteroaggregative E. coli (EAEC); (ii) enteroinvasive E. higher or lower level based on whether they were educated
coli (EIEC); (iii) enterohemorrhagic E. coli (EHEC); (iv) enter- (persons finishing at least college or university) or workers,
opathogenic E. coli (EPEC); (v) diffusely adherent E. coli farmers, and laborers (persons educated up to high school).

(DAEC); and (vi) enterotoxigenic E. coli (ETEC). It has been The living standard of the child’s family was evaluated by
shown that there are important regional differences in the monthly income of the whole family in Vietnamese Dong
prevalence of the different categories of DEC.12—15 (VND). Five levels (very poor, poor, middle, fair, and rich)
were ranked according to the Survey of the Center of Scien-
Besides group A rotavirus and DEC, the expanding list of tific Research for the Family and Woman carried out in
potential enteropathogens includes Salmonella spp, Shigella Vietnam in 2001. Water sources were divided into hygienic
spp, Vibrio cholerae, enterotoxigenic Bacteroides fragilis (piped water) and unhygienic (pool or well, or rainy water)
(ETBF), Campylobacter spp and Cryptosporidium spp. resources. A latrine was considered to be a hygienic conve-
Advances in diagnostic techniques have increased our ability nience. The availability of information on health and sanita-
to detect these pathogens. The present study was under- tion from any source was assessed according to whether the
taken with the aim of assessing the role of the enteric child’s family had access to this kind of information often
pathogens in relation to clinical symptoms and epidemiolo- (daily and weekly) or less often (monthly, rarely, or almost
gical factors. never).

Study subjects Materials and methods

A total of 836 children from 0 to 60 months of age including Sample collection
587 children with diarrhea attending the examination rooms
of three different hospitals and 249 age-matched healthy Fecal samples (one from each subject) from children without
controls were studied. The healthy children were enrolled diarrhea were collected in a clean container by their
from a daycare center and a healthcare center in Hanoi, parents when the children defecated. From the children with

300 T. Vu Nguyen et al.

diarrhea, one stool specimen was collected within 24 hours the Mann—Whitney U test (for nonparametric data) was used
of admission. All feces were collected in special containers for comparing two groups. A p value of <0.05 was considered
with Cary—Blair transport medium, kept at 4 8C and trans- statistically significant. Data from antibiotic susceptibility
ported to the microbiological laboratory within 24 hours. The testing were analyzed by WHONET 5.1 software.
residue of each sample after the first culture on media was
kept at À70 8C for further work. The collection of samples Ethical committee approval

stopped for two weeks for the TET holidays in Vietnam in
February 2002. Both the Ethical Committees at the Karolinska Institutet,
Sweden and Hanoi Medical University, Vietnam approved the
Isolation and identification of diarrheal project.
pathogens
Results
The methods for isolation and identification of the diarrheal
pathogens: rotavirus A, diarrheagenic E. coli, Shigella spp, Etiology and clinical properties of diarrhea
Salmonella spp, Vibrio cholerae and enterotoxigenic Bacter-
oides fragilis, have been published in detail previously and The rates of identification of different enteric pathogens are
are briefly described below.18—20 shown in Table 1.

For group A rotavirus: stool samples were analyzed for In 587 children with diarrhea, group A rotavirus was the
rotavirus A by using the IDEA rotavirus enzyme-linked immu- most frequently identified enteric pathogen with a preva-
nosorbent assay kit (DAKO Ltd, Ely, UK), according to the lence of 46.7%, showing a significant difference compared to
instructions of the manufacturer. This test is a qualitative the controls (3.6%). Within the diarrhea group, the detection
enzyme immunoassay for the detection of rotavirus (group A) prevalence in children less than two years of age was 51.1%,
in human fecal samples. significantly different ( p < 0.001) from that in the older
children (35.9%). Rotavirus infection was most prevalent in
For diarrheagenic E. coli, Shigella spp, Salmonella spp, children in the 13—24 months group. The second highest
Vibrio cholerae: stool samples were cultured on the surface number of cases were seen in the 0—12 months and 25—36
of (i) sorbitol MacConkey agar (Labora, Stockholm, Sweden) months age groups, although cases were also seen in the
for the selection of Escherichia coli isolates; (ii) thiosulfate older children. There was a significantly decreasing trend in
citrate bile salt cholera medium (Labora) for the selection of rotavirus prevalence with age (Chi-square test for trend,
Vibrio species; and (iii) deoxycholate citrate agar (Sigma- 8.904; p < 0.005).
Aldrich, Stockholm, Sweden) for the selection of Shigella spp
and Salmonella spp. The cultures were incubated overnight The second most common pathogen in the diarrhea group
at 37 8C. All samples were tested for Vibrio spp, Shigella spp, was DEC. The isolation prevalence was 22.5%. This included
and Salmonella spp by using colony morphology, biochemical 68 samples (11.6%) with EAEC, 12 (2.0%) with EIEC, 39 (6.6%)
properties, and agglutination with specific sera. A multiplex with EPEC, and 13 (2.2%) with ETEC. DEC accounted for 12% in

PCR using eight primer pairs specific for the virulent genes of the controls ( p < 0.001). The distribution was: 18 (7.2%) with
five different pathotypes of diarrheagenic E. coli was used for EAEC, 11 (4.4%) with EPEC, and one (0.4%) with ETEC. Of the
the identification of these E. coli. isolated ETEC in both groups, 7/14 (50%) produced heat-
labile toxins (LT) only, 4/14 (28.6%) LT and heat-stable
Enterotoxigenic Bacteroides fragilis was identified by toxins (ST), and 3/14 (21.4%) ST only. All the isolated EPEC
immunoseparation in combination with PCR. In brief, a were atypical. In the diarrhea group, EAEC and EPEC were
fecal sample suspension was inoculated into fastidious more frequently isolated in children less than two years of
anaerobe broth (FAB) medium (LAB 71, LAB M, Interna- age (14.1% and 7.9%, respectively), whereas EIEC and ETEC
tional Diagnostic Group, Bury, UK) and then incubated at were less frequently found (1.9% and 1%, respectively). No
37 8C for 48 h. After incubation, the broth medium was children were colonized with more than two DEC.
centrifuged twice. The pellet was suspended and incu-
bated with magnetic beads coated with monoclonal anti- Twenty-eight Shigella strains (4.7% of the samples) includ-
body (mAb C3) that binds specifically to a common epitope ing one S. boydii, seven S. flexneri, and 20 S. sonnei were
present in the inner core region of B. fragilis lipopolysac- isolated in the diarrhea group. The isolation prevalence of S.
charide (LPS). Bacteria bound to the coated beads were sonnei in children less than two years of age and the older
separated by magnetic separator. PCR using primers spe- ones was 1.4% and 8.2%, respectively. The difference was
cific for enterotoxin gene and its subtypes were applied for statistically significant ( p < 0.0001). Shigella ssp were not
ETBF identification. found in the healthy controls.

Campylobacter spp, and parasites were not investigated Within the group of children with diarrhea, 7.3% ETBF was
due to the lack of facilities in Hanoi, Vietnam. The methods detected. The corresponding figure for the controls was 2.4%
for antimicrobial susceptibility of the isolated E. coli and ( p < 0.01). Within the diarrhea group, the prevalence was
Shigella strains have been published elsewhere.21 significantly higher in children older than one year. Three
subtypes of ETBF isolates have been identified with preva-
Statistical analysis lences of 67.4%, 18.6%, and 16% for bft-1, bft-2, and bft-3,
respectively. In the controls, two of the subtypes were
The proportion difference was determined by the Chi-square identified, five bft-1 and one bft-2.
test. In the case where the expected value for a cell was <5,
Fisher’s exact test was used. Multiple comparisons of mean No EHEC, Salmonella spp, or V. cholerae were identified.
values of groups were done by the Kruskal—Wallis H test, and The occurrences of single and mixed infections of enteric

Diarrhea in children in Hanoi, Vietnam 301

Table 1 Distribution of identified pathogens according to age group in children with diarrhea and healthy controls

Identified pathogens Age group (months) Group of children No. (%) a p Value

Diarrhea (n = 587) Control (n = 249)

Group A rotavirus 0—12 111 (46.3) 1 (2.1) <0.00001
13—24 102 (57.6) 4 (5.6) <0.00001
25—36 3 (6.3) <0.00001
37—48 42 (44.2) 1 (2.0) <0.001
49—60 12 (29.3) 0 (0) <0.05

7 (20.6)

Diarrheagenic E. coli 0—12 35 (14.6) 2 (4.2) <0.05
EAEC 13—24 24 (13.6) 5 (7.0) >0.05
25—36 7 (14.6) >0.05
37—48 6 (6.3) 3 (6.1) >0.05
49—60 2 (4.9) 1 (3.0) >0.05
1 (2.9)

EIEC 0—12 2 (0.8) 0 (0) >0.05
6 (3.4) 0 (0) >0.05
13—24 2 (2.1) 0 (0) >0.05
0 (0) 0 (0) See b
25—36 2 (5.9) 0 (0) >0.05

37—48

49—60

EPEC 0—12 15 (6.3) 1 (2.1) >0.05
13—24 18 (10.2) 5 (7.0) >0.05
25—36 2 (4.2) >0.05
37—48 3 (3.2) 2 (4.1) >0.05
49—60 2 (4.9) 1 (3.0) >0.05
1 (2.9)

ETEC 0—12 2 (0.8) 0 (0) >0.05
13—24 2 (1.1) 1 (1.4) >0.05
25—36 4 (4.2) 0 (0) >0.05
37—48 3 (7.3) 0 (0) >0.05
49—60 2 (5.9) 0 (0) >0.05

Shigella spp 0—12 0 (0) 0 (0) See b
S. flexneri 13—24 3 (1.7) 0 (0) >0.05
25—36 1 (1.1) 0 (0) >0.05
37—48 3 (7.3) 0 (0) >0.05
49—60 0 (0) 0 (0) See b

S. sonnei 0—12 1 (0.4) 0 (0) >0.05
13—34 5 (2.8) 0 (0) >0.05
25—36 5 (5.3) 0 (0) >0.05
37—48 6 (14.6) 0 (0) <0.01
49—60 3 (8.8) 0 (0) >0.05

S. boydii 0—12 1 (0.4) 0 (0) See b

ETBF 0—12 12 (5.0) 0 (0) >0.05
13—34 16 (9.0) 3 (4.2) >0.05
25—36 1 (2.1) >0.05
37—48 5 (5.3) 2 (4.1) >0.05
49—60 5 (12.2) 0 (0) =0.053
5 (14.7)

a Percentage calculated according to total number in each age group.
b Could not perform statistical test.

pathogens are shown in Tables 2 and 3. Among children with The seasonality of infection was analyzed for rotavirus,
diarrhea, 79 (13.5%) had infections with two or more patho- DEC, Shigella spp, and ETBF. Rotavirus infection occurred
gens, and among control children, two (0.8%) had a mixed year-round but the prevalence trend was higher in Septem-
infection ( p < 0.00001). A potential enteric pathogen was ber—December, the cooler autumn and winter months. Infec-
identified from 395 children with diarrhea (67.3%) and 43 tions with other pathogens peaked during the summer time
controls (17.3%) ( p < 0.00001). when it was warm and rainy (Figure 1).

302 T. Vu Nguyen et al.

Table 2 Occurrence of single and mixed infections of enteric pathogens Total (n = 836)

Patterns of infection Group of children No. (%) 208 (24.9)
54 (6.5)
Diarrhea (n = 587) Control (n = 249) 10 (1.2)
29 (3.5)
Group A rotavirus 201 (34.2) 7 (2.8) 9 (1.1)
EAEC 37 (6.3) 17 (6.8) 5 (0.6)
EIEC 10 (1.7) 16 (1.9)
EPEC 19 (3.2) 0 (0) 1 (0.1)

ETEC 8 (1.4) 10 (4.0) 25 (3.0)
S. flexneri 5 (0.9) 30 (3.6)
S. sonnei 16 (2.7) 1 (0.4) 20 (2.4)
S. boydii 1 (0.2) 0 (0) 5 (0.6)
ETBF 19 (3.2) 0 (0) 1 (0.1)
Group A rotavirus + EAEC 29 (4.9) 0 (0) 1 (0.1)
Group A rotavirus + EPEC 19 (3.2) 6 (2.4) 15 (1.8)
Group A rotavirus + ETEC 5 (0.9) 1 (0.4) 2 (0.2)
Group A rotavirus + S. flexneri 1 (0.2) 1 (0.4) 1 (0.1)
Group A rotavirus + S. sonnei 1 (0.2) 0 (0) 3 (0.4)
Group A rotavirus + ETBF 15 (2.6) 0 (0) 2 (0.2)
ETBF + EIEC 2 (0.3) 0 (0) 1 (0.1)
ETBF + S. flexneri 1 (0.2) 0 (0)
ETBF + S. sonnei 3 (0.5) 0 (0) 398 (47.6)
Group A rotavirus + EAEC + ETBF 2 (0.3) 0 (0)
Group A rotavirus + EPEC + ETBF 1 (0.2) 0 (0)
No pathogen 0 (0)
192 (32.7) 0 (0)
206 (82.7)

Table 3 Infection with single and mixed pathogens in children with diarrhea and controls

No. of infecting pathogens Group of children No. (%) Total (n = 836)

Diarrhea (n = 587) Control (n = 249) 398
357
No pathogen 192 (32.7) 206 (82.7)
One pathogen 316 (53.8) 41 (16.5) 78
Two pathogens 2 (0.8) 3
Three pathogens 76 (13.0) 0 (0)

3 (0.5)

The susceptibility of isolated DEC and Shigella strains was and fever. The different categories accounted for 54.5, 19.3,
tested against eight antibiotics. Imipenem (IPM), ciproflox- 16.7, and 9.5%, respectively. Overall, when being examined,
acin (CIP), nalidixic acid (NAL), cefotaxime (CTX), and cefur- 43.6% of children with diarrhea were febrile, 53.8% had
oxime (CXM) were active against the E. coli pathogens, while vomiting, and 82.6% had dehydration. Types of diarrheal
high frequencies of resistance to ampicillin (AMP), chloram-
phenicol (CHL), and trimethoprim/sulfamethoxazole (SXT) Figure 1 Seasonal prevalence of the identified enteric patho-
were shown, with resistance prevalences of 86.4%, 77.2%, gens in children less than five years of age in Hanoi, Vietnam.
and 88.3%, respectively. Nearly 89% of the Shigella strains Rotavirus (&); Escherichia coli (^); Shigella spp (*) and enter-
were resistant to trimethoprim/sulfamethoxazole, 75% were otoxigenic Bacteroides fragilis (~).
resistant to ampicillin, and 53.6% were resistant to chlor-
amphenicol. More than 85% of the strains were susceptible to
cefuroxime, cefotaxime, nalidixic acid, ciprofloxacin, and
imipenem. Multi-antibiotic resistance was detected in 145/
162 (89.5%) of the diarrheagenic E. coli and 22/28 (78.6%) of
the Shigella strains.

The most prevalent multiresistance patterns (the resis-
tance to at least two antibiotics) for all E. coli and Shigella
strains were AMPr CHLr CXMs CTXs NALs CIPs IPMs SXTr and AMPr
CHLs CXMs CTXs NALs CIPs IPMs SXTr in 89.5% of all E. coli and in
35% of Shigella strains, respectively.

In the diarrhea group the reasons for visiting the hospital
were: (i) diarrhea only; (ii) diarrhea and vomiting; (iii)
diarrhea and fever; and (iv) diarrhea together with vomiting

Diarrhea in children in Hanoi, Vietnam 303

stool were noted. Watery stool was predominant with a for the diarrhea group and 1.18 for the control group showing
prevalence of 66.4% followed by mucous stool (21%). There a significant difference (Table 4). All were less than five years
was one bloody stool (0.2%). Other types of stool of age. The age distribution of all subjects is shown in Table 5.
accounted for 12.4%. The mean episodes of stools per Of those with diarrhea, 40.9% were less than one year old and
day was seven, ranging from 2 to 23. Before attending 71.0% were less than two years old. Table 4 shows some
the hospital, 162/587 (27.6%) of children had been given characteristics of the children in both groups. The children
antibiotics and 523/587 (89.1%) of the children received without diarrhea had a current average weight significantly
oral rehydration fluid. higher than those with diarrhea both in children 2 years old
and in those >2 years old. Of patients less than six months of
Epidemiology of diarrhea age, 22.3% from the diarrhea group were fully breast-fed as
compared to 36.4% of the controls. The difference is not
During the period from March 2001 to April 2002, we studied statistically significant. However, the corresponding figures
836 subjects living in Hanoi including 587 children with in children up to three months of age were 36.8% and 71.4%,
diarrhea and 249 controls. The male/female ratio was 1.64 respectively ( p = 0.026).

Table 4 Epidemiological factors related to the risk for diarrhea

Characteristic Group of children p Value

Diarrhea (n = 587) Control (n = 249) <0.05
135/114
Child characteristics 365/222 <0.05
Gender male/female: 3.16 >0.05
Weight 3.06 3.16 <0.01
Mean weight at birth (kg) 3.10 9.61 <0.01
Children 2 years old 14.92 >0.05
Children >2 years old 8.50 12 (36.4%) a
Mean of current weight (kg) 13.66 <0.0001
Children 2 years old 23 (22.3%) a 14 (5.6%) <0.0001
Children >2 years old 128 (51.4%) <0.001

211 (84.7%) <0.0001
Only breast-fed (children 6 months) 231 (92.8%) <0.0001
247 (99.2%) Chi-square
Clinical symptoms (diarrhea only) 256 (43.6%) for
Fever 316 (53.8%) 1 (0.4%) trend
Vomiting 485 (82.6%) 58 (23.3%) =3.403
Dehydration 90 (36.1%) p = 0.065
Kinds of stool 390 (66.4%) 53 (21.3%) <0.05
Watery 1 (0.2%) 47 (18.9%)
Bloody 65 (26.1%)
Mucous 123 (21.0%)
Other 73 (12.4%)
Stools per dayb (mean) 7

Mother characteristics 102 (17.4%)
Age <25 206 (35.1%)
High education level

Hygiene conditions 250 (42.6%)
Hand washing c 408 (69.5%)
Water resource (piped water) 469 (79.9%)
Convenience (hygienic latrine)

Living standard 18 (3.1%)
Very poor 152 (25.9%)
Poor 204 (34.7%)
Middle 121 (20.6%)
Fair
Rich 92 (15.7%)

Information on health and sanitation 112 (19.1%)
Daily, weekly

a Percentage in the defined group.
b Episodes per day when children were examined.
c Mother’s hand washing before feeding children.

304 T. Vu Nguyen et al.

Table 5 Distribution of children by age group countries. The epidemiological significance of each E. coli
category in childhood diarrhea varies with geographical area.
Age group No. (%) of children Total
(months) ETEC causes a significant number of cases of childhood
Diarrhea Control 288 (34.4) diarrhea and gastroenteritis among travelers. In our study,
248 (29.7) we did not see any strong association of ETEC with diarrhea as
0—12 240 (40.9) 48 (19.3) 143 (17.1) compared to controls ( p = 0.076). However, the prevalence
13—24 177 (30.1) 71 (28.5) of ETEC was significantly higher in children over two years of
25—36 48 (19.3) 90 (10.8) age as compared to those less than two years of age in the
37—48 95 (16.2) 49 (19.7) 67 (8.0) diarrhea group (Figure 2) ( p < 0.005). LT+ ETEC, ST+ ETEC,
49—60 41 (7.0) 33 (13.2) and LT+ ST+ ETEC strains were isolated with a lower preva-
34 (5.8) lence, 2.2% in patients, as compared to the study by Wolk
et al.35 but the same prevalence as in the study by Nishi
Table 4 shows data concerning other epidemiological et al.36 As described in previous studies, the prevalence of
factors related to demography and the socio-economic situa- ETEC is different in different geographical areas. According
tion of the children’s families. A significantly higher preva- to Mayatepek et al., the figures in the diarrhea group and in
lence of mothers aged under 25 was seen in children with the control group were 28% and 16%, respectively;37 Wolk
diarrhea compared to the controls. With regard to living et al. showed the prevalence of ETEC to be 20.7% in
standards, there was a decreasing trend in children with patients.35
diarrhea with an increasing level of family income
( p = 0.065). Overall, the prevalence of children with diar- EHEC is an important food-borne pathogen, especially in

rhea was significantly higher in families where hygienic water developed countries. Clinical manifestations of EHEC infec-
and latrines were lacking, where mothers less often washed tion range from asymptomatic carriage to diarrhea to hemor-
their hands before feeding the children, where mothers had a rhagic colitis. Hemolytic uremic syndrome (HUS) is a common
low level of education, and where information on health and complication in children.38 No EHEC was isolated in this
sanitation reached the family less often. study. This is not uncommon and similar results have been
obtained by others.27,39,40
Discussion
EPEC continues to be an important cause of diarrhea in
Etiology and clinical properties of diarrhea children up to two years of age.41 Our study showed a slightly
higher prevalence of EPEC in patients compared to controls.
Diarrhea in developing countries is caused by an increas- It seems to be associated with children under 2 years of age
ingly long list of viral, bacterial, and parasitic pathogens ( p = 0.053) as compared to the older ones. All of the isolated
with rotavirus, diarrheagenic E. coli, Shigella spp, Salmo- EPEC were atypical strains having only the eaeA gene but no
nella spp, and V. cholerae heading the list.22 In the present bfpA, which is the structural gene encoding BFP (bundle-
study, no Salmonella or V. cholerae strains were found. The forming pilus). Since bfpA is encoded on the EAF plasmid42 it
reason for this could be the low prevalence of these patho- could be lost during culture and isolation, resulting in the
gens as shown in previous studies in Vietnam.23,24 However, failure to identify them in our study.
in other investigations in developing countries,16,25—27
these agents have been identified more often in fecal In other studies, children colonized with EPEC were rarely
samples from children with diarrhea. The other pathogens: infected with other enteric pathogens.10,43—45 In this study,
group A rotavirus, DEC, Shigella spp, and ETBF were iden- 20 of the EPEC-positive children (19 from the patients
tified with prevalences of 46.7, 22.5, 4.8, and 7.3%, respec- and one from the controls) were also positive for group A
tively. A recognized pathogen was identified in 67.3%
patients. This is a rather high prevalence as compared to Figure 2 Comparison of the prevalence of diarrheagenic
other studies, even in adult patients,12,16,28,29 but still Escherichia coli in the diarrhea group in children younger (&)
lower than the data of Albert et al.26 with a potential and older (&) than two years of age. Significant differences are
enteric pathogen in 74.8%. seen in EAEC ( p < 0.005) and ETEC ( p < 0.005).

Rotavirus is the most important cause of diarrhea in
children in developing countries, causing around one million

deaths per year.30 Several studies in Vietnam have shown that
group A rotavirus plays an important role in diarrheal disease
in children.31—34 In this study we have confirmed the high
prevalence of rotavirus infection in Vietnamese children with
diarrhea, showing a significant difference compared to con-
trols. Moreover, information concerning clinical properties
and co-infections of rotavirus has been investigated and this
has contributed to the knowledge of diarrhea caused by
rotavirus in children in Vietnam. A limitation with respect
to rotavirus was that we could not identify the serotypes of
this pathogen.

Diarrheagenic E. coli are recognized as an emerging etiol-
ogy causing diarrhea in children, especially in developing

Diarrhea in children in Hanoi, Vietnam 305

rotavirus, and one harbored group A rotavirus, EPEC, and floxacin) in the treatment of diarrhea caused by DEC and
ETBF, simultaneously (Table 2). The co-infection of EPEC with Shigella spp should also be carefully considered since many
other pathogens has also been described in another study.26 strains have developed resistance to these agents as shown in
previous studies.69—72 To our knowledge, there have not been
Regarding EIEC infection, several previous studies27,46,47 any published studies on antibiotic resistance of all five
have found no or very few EIEC from fecal samples in children different categories of DEC in Vietnam, so far.
with diarrhea. Only 2% of EIEC were isolated from patients
in our study. It has been shown that in some developing Epidemiology of diarrhea
countries, EIEC strains can be responsible for 1 to 7% of cases
of diarrhea or dysentery.12,48—50 EIEC and Shigella spp are The epidemiology of enteric pathogens that cause diarrhea
both responsible for dysentery diarrhea. Although Shigella suggests that most infections are acquired from food, water,
spp were isolated with a higher prevalence, it was still lower and hand contact and many diarrheal diseases can be pre-
than that of other studies.26,49,51 vented by simple rules of personal hygiene and safer food

preparation.73,74 Among the contributors to diarrhea in chil-
EAEC was the most predominant bacterial pathogen in this dren, the household health environment, living standards,
study for both children with and without diarrhea. It was and mothers’ knowledge play important roles. Women play a
strongly associated with diarrhea in children less than two major role in making or breaking the ‘chain of contamination’
years of age. Many surveillance studies have demonstrated within the household sphere.75 Their knowledge of care-
the importance of EAEC in pediatric diarrhea.15,42,52,53 How- giving for their children with diarrhea is very important. In
ever, there have been conflicting reports on the association the present study, 89.1% of children with diarrhea had been
of EAEC with acute and persistent diarrhea.15,25,27,54,55 EAEC given oral rehydration fluid before hospitalization. Seventy
was not associated with persistent diarrhea in this study. It percent of them had received oral rehydration solutions
could be due to the largely unknown or different epidemio- (ORS). In an investigation in Vietnam,76 60% of mothers knew
logical characteristics of EAEC (e.g., likely sources, reser- about ORS and 40% of children with diarrhea were given ORS.
voirs of infection, routes of transmission, seasonality, and However, the awareness is different in different age groups,
age-distribution). the lowest level of awareness being in the under 25s. Educa-
tion level also plays an important role in the knowledge of
Very few studies on DEC have been carried out in Viet- mothers. It has been shown that the more education they
nam.23,24 The findings concerning this emerging pathogen have, the more they know about the way to take care of
will be helpful for pediatricians and microbiologists in diag- children with diarrhea, and the lower prevalence of diarrhea
nosis and treatment of children with diarrhea. their children may have. This study, along with others, has
shown that diarrhea is more common among children whose
ETBF has been studied intensively for the last 20 years mothers are in the younger age group, and have a low level of
since it was recognized as an important anaerobic bacterial education.8,77,78
pathogen causing diarrhea in small children.56—58 This is the
first study of the role of ETBF in children’s diarrhea in Diarrheal diseases are water-, hygiene-, food-, and sani-
Vietnam and it is concluded that this pathogen is an impor- tation-related and have multiple oral—fecal transmission
tant causative agent of diarrhea in children in Hanoi, Viet- routes.79—81 In this study, significantly higher prevalences
nam. of families who had piped water, a hygienic latrine, and
where the mothers more often washed their hands before
There were no striking differences regarding vomiting, feeding their children were seen in the group of healthy
dehydration, and episode/day diarrhea in children infected children as compared to the group of children with diarrhea.
with different categories of DEC and Shigella. Children

infected with Shigella had fever with a temperature average Hand washing can interrupt some transmission routes of
significantly higher than that of those infected with rota- enteric pathogens to the host. There are a number of epi-
virus, DEC, and ETBF. In addition, the average age of these demiological studies on hand washing which claim substantial
children was also significantly higher. Nevertheless, vomiting reduction in diarrheal morbidity.82—85 However, water avail-
and watery diarrhea were strongly associated with rotavirus ability is likely to have an impact on the frequency of hand
infection. Children infected with rotavirus alone had a sig- washing.
nificantly higher prevalence of vomiting and watery diarrhea
as compared to those infected with other pathogens It is not easy for every family to get a hygienic water
( p < 0.05 and p < 0.05, respectively). These can be useful supply for domestic use. It depends on the geographical area,
symptoms in clinical diagnosis and examination of rotavirus supply infrastructure, and the supply capacity of the water
diarrhea as mentioned in previous studies.59—62 plant. In Hanoi, Vietnam, people in some areas still do not
have access to sufficient piped water. During the summer,
The peak of rotavirus infection occurred during the winter when water consumption rises dramatically, some families
months, and the bacterial infections were often predominant face a shortage of hygienic water. They may have to find
during the summer or warm months. These trends agree with additional sources or try to store water in different ways.
the findings of other studies.10,63—65 These facilitate microbial contamination resulting in diar-
rhea, especially in children. This could partially explain the
There is an increasing concern for the possible develop- high prevalence of diarrhea in children during the summer
ment of resistance to antimicrobial agents in the Enterobac- months.
teriaceae, especially in DEC when this pathogen has been
emerging, and the resistance is a result of inappropriate use As mentioned above, there is an association between stool
of antibiotics in hospitals and communities.66—68 Our data disposal and child diarrhea. We have seen that diarrhea could
show that ampicillin, chloramphenicol and trimethoprim/ be prevented in about 20% of the children by having access to
sulfamethoxazole should not be considered as appropriate
for empirical therapy of children with diarrhea in Vietnam.
Moreover, the use of quinolones (nalidixic acid and cipro-

306 T. Vu Nguyen et al.

hygienic latrines in the households. In fact, not all the house- 6. Ahiadeke C. Breast-feeding, diarrhea and sanitation as compo-

holds in Hanoi have their own hygienic latrines. In some cases nents of infant and child health: a study of large scale survey
they have to share with others. Sometimes these types of data from Ghana and Nigeria. J Biosoc Sci 2000;32:47—61.
conveniences are not hygienic. People in some areas use an
indigenous composting latrine or defecation pit. The problem 7. Magnani RJ, Mock NB, Bertrand WE, Clay DC. Breast-feeding,
of safe stool disposal has also been investigated in many other water and sanitation, and childhood malnutrition in the Philip-
studies.86—89 Safe stool disposal is one of the key barriers to pines. J Biosoc Sci 1993;25:195—211.
the transmission of enteric pathogens.
8. Perera BJ, Ganesan S, Jayarasa J, Ranaweera S. The impact of
Diarrhea prevention is an important public health issue breastfeeding practices on respiratory and diarrheal disease in
and it is necessary to disseminate information and teach infancy: a study from Sri Lanka. J Trop Pediatr 1999;45:115—8.
skills, and appropriate activities to the public. However,
the availability of this type of information and delivery to 9. Parashar UD, Bresee JS, Gentsch JR, Glass RI. Rotavirus. Emerg
the households, plays an important role in the anti-diarrhea Infect Dis 1998;4:561—70.
campaign. In our study, to the question: ‘‘How often does
your family receive information about health and sanita- 10. Gomes TA, Rassi V, MacDonald KL, Ramos SR, Trabulsi LR, Vieira
tion?’’, 26.1% of households in the healthy group claimed MA, et al. Enteropathogens associated with acute diarrheal
that they received the information daily or weekly, compared disease in urban infants in Sao Paulo, Brazil. J Infect Dis
to 19.1% in the diarrhea group, showing a significant differ- 1991;164:331—7.
ence. This shows that, although many diarrheal diseases can
be prevented by following the simple rules of personal 11. de Zoysa I, Feachem RG. Interventions for the control of diar-
hygiene and safe food preparation, public healthcare workers rheal diseases among young children: rotavirus and cholera
should pay more attention to effectively inform the popula- immunization. Bull World Health Organ 1985;63:569—83.
tion about health and sanitation through mass media to the
households. 12. Echeverria P, Taylor DN, Lexsomboon U, Bhaibulaya M, Blacklow
NR, Tamura K, et al. Case-control study of endemic diarrheal
In conclusion, many aspects of diarrhea need to be further disease in Thai children. J Infect Dis 1989;159:543—8.
investigated in order to decrease the prevalence of diarrhea
in Vietnamese children under five years of age. 13. Cravioto A, Reyes RE, Ortega R, Fernandez G, Hernandez R,
Lopez D. Prospective study of diarrheal disease in a cohort of
Acknowledgments rural Mexican children: incidence and isolated pathogens during

the first two years of life. Epidemiol Infect 1988;101:123—34.
This work was supported by the Swedish Agency for Research
Cooperation with Developing Countries SIDA/SAREC. The 14. Bouzari S, Jafari A, Farhoudi-Moghaddam AA, Shokouhi F, Parsi M.
authors thank the technicians, nurses, and staff at the Divi- Adherence of non-enteropathogenic Escherichia coli to HeLa
sion of Clinical Bacteriology, Substrate Department, Karo- cells. J Med Microbiol 1994;40:95—7.
linska University Hospital, Huddinge, Stockholm, Sweden; at
the Department of Medical Microbiology, Hanoi Medical Uni- 15. Bhan MK, Raj P, Levine MM, Kaper JB, Bhandari N, Srivastava R,
versity; at the Department of Gastroenterology, Vietnam— et al. Enteroaggregative Escherichia coli associated with persis-
Sweden Hospital, Hanoi; at the Kim Lien Day Care Centre; and tent diarrhea in a cohort of rural children in India. J Infect Dis
at the Health Care Centre, Dong Da, Hanoi, Vietnam, for their 1989;159:1061—4.
devoted microbiological expertise and assistance with col-
lection, transportation, analyses of samples and technical 16. Baqui AH, Sack RB, Black RE, Haider K, Hossain A, Alim AR, et al.
guidance for lab work. Enteropathogens associated with acute and persistent diarrhea
in Bangladeshi children less than 5 years of age. J Infect Dis
Conflict of interest: No conflict of interest to declare. 1992;166:792—6.

References 17. Anonymous. Reading on diarrhea, Student Manual. World Health
Organization 1992.
1. Rodrigues J, Acosta VC, Candeias JM, Souza LO, Filho FJ. Pre-
valence of diarrheogenic Escherichia coli and rotavirus among 18. Nguyen TV, Van PL, Gia KN, Huy CL, Weintraub A. Detection and
children from Botucatu, Sao Paulo State, Brazil. Braz J Med Biol characterization of diarrheagenic E. coli from children under 5
Res 2002;35:1311—8. years of age in Hanoi, Vietnam. J Clin Microbiol 2005;43:755—
60.
2. Tulloch J, Richards L. Childhood diarrhea and acute respiratory
infections in developing countries. Med J Aust 1993;159: 19. Nguyen TV, Van PL, Huy CL, Weintraub A. Diarrhea caused by
46—51. enterotoxigenic Bacteroides fragilis in children less than five
years of age in Hanoi, Vietnam. Anaerobe 2005;11:109—14.
3. Guilbert JJ. The world health report 2002 — reducing risks,
promoting healthy life. Educ Health (Abingdon) 2003;16: 20. Nguyen TV, Van PL, Huy CL, Weintraub A. Diarrhea caused by
230. rotavirus in children less than five years of age in Hanoi, Vietnam.

J Clin Microbiol 2004;42:5745—50.
4. Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJ.
Selected major risk factors and global and regional burden of 21. Nguyen TV, Van PL, Huy CL, Weintraub A. Antibiotic resistance in
disease. Lancet 2002;360:1347—60. diarrheagenic Escherichia coli and Shigella strains isolated from
children in Hanoi, Vietnam. Antimicrob Agents Chemother
5. Bhandari N, Bahl R, Sazawal S, Bhan MK. Breast-feeding status 2005;49:816—9.
alters the effect of vitamin A treatment during acute diarrhea in
children. J Nutr 1997;127:59—63. 22. Taylor DN, Echeverria P. Diarrheal disease: current concepts and
future challenges. Molecular biological approaches to the epi-
demiology of diarrheal diseases in developing countries. Trans R
Soc Trop Med Hyg 1993;87(Suppl 3):3—5.

23. Isenbarger DW, Hien BT, Ha HT, Ha TT, Bodhidatta L, Pang LW,
et al. Prospective study of the incidence of diarrhea and pre-
valence of bacterial pathogens in a cohort of Vietnamese chil-
dren along the Red River. Epidemiol Infect 2001;127:229—36.

24. Ngan PK, Khanh NG, Tuong CV, Quy PP, Anh DN, Thuy HT.
Persistent diarrhea in Vietnamese children: a preliminary report.
Acta Paediatr Suppl 1992;381:124—6.

25. Arthur JD, Bodhidatta L, Echeverria P, Phuphaisan S, Paul S.
Diarrheal disease in Cambodian children at a camp in Thailand.
Am J Epidemiol 1992;135:541—51.

26. Albert MJ, Faruque AS, Faruque SM, Sack RB, Mahalanabis D.
Case-control study of enteropathogens associated with child-

Diarrhea in children in Hanoi, Vietnam 307

hood diarrhea in Dhaka, Bangladesh. J Clin Microbiol 1999; EPEC adherence factor genes in infants with and without diar-
37:3458—64. rhea in Sao Paulo, Brazil. J Infect Dis 1989;160:131—5.
27. Albert MJ, Faruque SM, Faruque AS, Neogi PK, Ansaruzzaman M, 46. Ogunsanya TI, Rotimi VO, Adenuga A. A study of the aetiological
Bhuiyan NA, et al. Controlled study of Escherichia coli diarrheal agents of childhood diarrhea in Lagos, Nigeria. J Med Microbiol
infections in Bangladeshi children. J Clin Microbiol 1995; 1994;40:10—4.
33:973—7. 47. Strockbine NA, Faruque SM, Kay BA, Haider K, Alam K, Alam AN,
28. Bourgeois AL, Gardiner CH, Thornton SA, Batchelor RA, Burr DH, et al. DNA probe analysis of diarrheagenic Escherichia coli:
Escamilla J, et al. Etiology of acute diarrhea among United detection of EAF-positive isolates of traditional enteropatho-
States military personnel deployed to South America and West genic E. coli serotypes among Bangladeshi paediatric diarrhea
Africa. Am J Trop Med Hyg 1993;48:243—8. patients. Mol Cell Probes 1992;6:93—9.
29. Stoll BJ, Glass RI, Huq MI, Khan MU, Holt JE, Banu H. Surveillance 48. Sunthadvanich R, Chiewsilp D, Seriwatana J, Sakazaki R, Eche-
of patients attending a diarrheal disease hospital in Bangladesh. verria P. Nationwide surveillance program to identify diarrhea-
Br Med J (Clin Res Ed) 1982;285:1185—8. causing Escherichia coli in children in Thailand. J Clin Microbiol
30. Bern C, Unicomb L, Gentsch JR, Banul N, Yunus M, Sack RB, et al. 1990;28:469—72.
Rotavirus diarrhea in Bangladeshi children: correlation of dis- 49. Echeverria P, Sethabutr O, Serichantalergs O, Lexomboon U,
ease severity with serotypes. J Clin Microbiol 1992;30:3234—8. Tamura K. Shigella and enteroinvasive Escherichia coli infections
31. Doan LT, Okitsu S, Nishio O, Pham DT, Nguyen DH, Ushijima H. in households of children with dysentery in Bangkok. J Infect Dis
Epidemiological features of rotavirus infection among hospita- 1992;165:144—1447.
lized children with gastroenteritis in Ho Chi Minh City, Vietnam. J 50. Faundez G, Figueroa G, Troncoso M, Cabello FC. Characteriza-
Med Virol 2003;69:588—94. tion of enteroinvasive Escherichia coli strains isolated from
32. Nguyen VM, Nguyen VT, Huynh PL, Dang DT, Nguyen TH, Phan VT, children with diarrhea in Chile. J Clin Microbiol 1988;26:928—
et al. The epidemiology and disease burden of rotavirus in 32.
Vietnam: sentinel surveillance at 6 hospitals. J Infect Dis 51. Ming ZF, Xi ZD, Dong CS, Serichantalergs O, Changchawalit S,
2001;183:1707—12. Nirdnoy W, et al. Diarrheal disease in children less than one year
33. Nishio O, Matsui K, Lan DT, Ushijima H, Isomura S. Rotavirus of age at a children’s hospital in Guangzhou, People’s Republic of
infection among infants with diarrhea in Vietnam. Pediatr Int China. Trans R Soc Trop Med Hyg 1991;85:667—9.
2000;42:422—4. 52. Kang G, Mathan MM, Mathan VI. Evaluation of a simplified HEp-2
34. Landaeta ME, Dove W, Vinh H, Cunliffe NA, Campbell J, Parry CM, cell adherence assay for Escherichia coli isolated from south
et al. Characterization of rotaviruses causing diarrhea in Viet- Indian children with acute diarrhea and controls. J Clin Microbiol
namese children. Ann Trop Med Parasitol 2003;97:53—9. 1995;33:2204—5.

35. Wolk M, Ohad E, Shafran R, Safir S, Cohen Y, Wiklund G, et al. 53. Paul M, Tsukamoto T, Ghosh AR, Bhattacharya SK, Manna B,
Epidemiological aspects of enterotoxigenic Escherichia coli diar- Chakrabarti S, et al. The significance of enteroaggregative
rhea in infants in the Jerusalem area. Public Health Rev Escherichia coli in the etiology of hospitalized diarrhea in Cal-
1995;23:25—33. cutta, India and the demonstration of a new honey-combed
36. Nishi J, Yoshinaga M, Manago K, Kuroki H, Kamenosono A, Mogi A, pattern of aggregative adherence. FEMS Microbiol Lett
et al. The incidences of virulence genes in Escherichia coli strains 1994;117:319—25.
from sporadic diarrheal children. Kansenshogaku Zasshi 1999; 54. Cravioto A, Tello A, Navarro A, Ruiz J, Villafan H, Uribe F, et al.
73:1104—9. Association of Escherichia coli HEp-2 adherence patterns with
37. Mayatepek E, Seebass E, Hingst V, Kroeger A, Sonntag HG. type and duration of diarrhea. Lancet 1991;337:262—4.
Prevalence of enteropathogenic and enterotoxigenic Escherichia 55. Fang GD, Lima AA, Martins CV, Nataro JP, Guerrant RL. Etiology
coli in children with and without diarrhea in Esteli, Nicaragua. J and epidemiology of persistent diarrhea in northeastern Brazil: a
Diarrheal Dis Res 1993;11:169—71. hospital-based, prospective, case-control study. J Pediatr Gas-
38. Kehl SC. Role of the laboratory in the diagnosis of enterohemor- troenterol Nutr 1995;21:137—44.
rhagic Escherichia coli infections. J Clin Microbiol 2002; 56. Kato N, Liu C, Kato H, Watanabe K, Nakamura H, Iwai N, et al.
40:2711—5. Prevalence of enterotoxigenic Bacteroides fragilis in children
39. Orlandi PP, Silva T, Magalhaes GF, Alves F, de Almeida Cunha RP, with diarrhea in Japan. J Clin Microbiol 1999;37:801—3.
Durlacher R, et al. Enteropathogens associated with diarrheal 57. Krzyzanowsky F, Avila-Campos MJ. Detection of non-enterotoxi-
disease in infants of poor urban areas of Porto Velho, Rondonia: a genic and enterotoxigenic Bacteroides fragilis in stool samples
preliminary study. Mem Inst Oswaldo Cruz 2001;96:621—5. from children in Sao Paulo, Brazil. Rev Inst Med Trop Sao Paulo
40. Torres ME, Pirez MC, Schelotto F, Varela G, Parodi V, Allende F, 2003;45:225—7.
et al. Etiology of children’s diarrhea in Montevideo, Uruguay: 58. Sack RB, Albert MJ, Alam K, Neogi PK, Akbar MS. Isolation of
associated pathogens and unusual isolates. J Clin Microbiol enterotoxigenic Bacteroides fragilis from Bangladeshi children
2001;39:2134—9. with diarrhea: a controlled study. J Clin Microbiol 1994;32:
41. Levine MM, Edelman R. Enteropathogenic Escherichia coli of 960—3.
classic serotypes associated with infant diarrhea: epidemiology 59. Cascio A, Vizzi E, Alaimo C, Arista S. Rotavirus gastroenteritis in
and pathogenesis. Epidemiol Rev 1984;6:31—51. Italian children: can severity of symptoms be related to the
42. Nataro JP, Steiner T, Guerrant RL. Enteroaggregative Escherichia infecting virus? Clin Infect Dis 2001;32:1126—32.
coli. Emerg Infect Dis 1998;4:251—61. 60. Barnes GL, Uren E, Stevens KB, Bishop RF. Etiology of acute
43. Echeverria P, Orskov F, Orskov I, Knutton S, Scheutz F, Brown JE, gastroenteritis in hospitalized children in Melbourne, Australia,
et al. Attaching and effacing enteropathogenic Escherichia coli from April 1980 to March 1993. J Clin Microbiol 1998;36:

as a cause of infantile diarrhea in Bangkok. J Infect Dis 133—8.
1991;164:550—4. 61. Bok K, Castagnaro N, Borsa A, Nates S, Espul C, Fay O, et al.
44. Gomes TA, Blake PA, Trabulsi LR. Prevalence of Escherichia coli Surveillance for rotavirus in Argentina. J Med Virol 2001;65:
strains with localized, diffuse, and aggregative adherence to 190—8.
HeLa cells in infants with diarrhea and matched controls. J Clin 62. Rytlewska M, Bako W, Ratajczak B, Marek A, Gwizdek A, Czar-
Microbiol 1989;27:266—9. necka-Rudnik D, et al. Epidemiological and clinical character-
45. Gomes TA, Vieira MA, Wachsmuth IK, Blake PA, Trabulsi LR. istics of rotaviral diarrhea in children from Gdansk, Gdynia and
Serotype-specific prevalence of Escherichia coli strains with Sopot. Med Sci Monit 2000;6:117—22.

308 T. Vu Nguyen et al.

63. Fang ZY, Yang H, Zhang J, Li YF, Hou AC, Ma L, et al. Child 75. Elmendorf M. Water supply, sanitation and diarrhea: the role of
rotavirus infection in association with acute gastroenteritis in women. Dialogue Diarrhea 1987;30:3.
two Chinese sentinel hospitals. Pediatr Int 2000;42:401—5.
76. Anonymous. Vietnam, Demographic and Health Survey 2002.
64. Maneekarn N, Ushijima H. Epidemiology of rotavirus infection in National Committee for Population, Family and Children 2003.
Thailand. Pediatr Int 2000;42:415—21.
77. Bani IA, Saeed AA, Othman AA. Diarrhea and child feeding
65. Levine MM, Ferreccio C, Prado V, Cayazzo M, Abrego P, Martinez practices in Saudi Arabia. Public Health Nutr 2002;5:727—31.
J, et al. Epidemiologic studies of Escherichia coli diarrheal
infections in a low socioeconomic level peri-urban community 78. Rasheed P. Perception of diarrheal diseases among mothers and
in Santiago, Chile. Am J Epidemiol 1993;138:849—69. mothers-to-be: implications for health education in Saudi Ara-
bia. Soc Sci Med 1993;36:373—7.
66. Bruinsma N, Filius PM, van den Bogaard AE, Nys S, Degener J,
Endtz HP, et al. Hospitalization, a risk factor for antibiotic- 79. Huttly SR, Morris SS, Pisani V. Prevention of diarrhea in young
resistant Escherichia coli in the community? J Antimicrob Che- children in developing countries. Bull World Health Organ
mother 2003;51:1029—32. 1997;75:163—74.

67. Chuc NT, Larsson M, Falkenberg T, Do NT, Binh NT, Tomson GB. 80. Esrey SA. Water, waste, and well-being: a multicountry study. Am
Management of childhood acute respiratory infections at private J Epidemiol 1996;143:608—23.

pharmacies in Vietnam. Ann Pharmacother 2001;35:1283—8.
81. Curtis V, Cairncross S, Yonli R. Domestic hygiene and diarrhea —
68. Cao V, Lambert T, Nhu DQ, Loan HK, Hoang NK, Arlet G, et al. pinpointing the problem. Trop Med Int Health 2000;5:22—32.
Distribution of extended-spectrum beta-lactamases in clinical
isolates of Enterobacteriaceae in Vietnam. Antimicrob Agents 82. Wilson JM, Chandler GN, Muslihatun, Jamiluddin. Hand-washing
Chemother 2002;46:3739—43. reduces diarrhea episodes: a study in Lombok, Indonesia. Trans R
Soc Trop Med Hyg 1991;85:819—21.
69. Zaidi MB, Zamora E, Diaz P, Tollefson L, Fedorka-Cray PJ, Head-
rick ML. Risk factors for fecal quinolone-resistant Escherichia 83. Pinfold JV, Horan NJ. Measuring the effect of a hygiene behaviour
coli in Mexican children. Antimicrob Agents Chemother intervention by indicators of behaviour and diarrheal disease.
2003;47:1999—2001. Trans R Soc Trop Med Hyg 1996;90:366—71.

70. Isenbarger DW, Hoge CW, Srijan A, Pitarangsi C, Vithayasai N, 84. Khan MU. Interruption of shigellosis by hand washing. Trans R Soc
Bodhidatta L, et al. Comparative antibiotic resistance of diar- Trop Med Hyg 1982;76:164—8.
rheal pathogens from Vietnam and Thailand, 1996—1999. Emerg
Infect Dis 2002;8:175—80. 85. Hoque BA, Juncker T, Sack RB, Ali M, Aziz KM. Sustainability of a
water, sanitation and hygiene education project in rural Bangla-
71. Hoge CW, Gambel JM, Srijan A, Pitarangsi C, Echeverria P. Trends desh: a 5-year follow-up. Bull World Health Organ 1996;74:
in antibiotic resistance among diarrheal pathogens isolated in 431—7.
Thailand over 15 years. Clin Infect Dis 1998;26:341—5.
86. Gorter AC, Sandiford P, Pauw J, Morales P, Perez RM, Alberts H.
72. Ashkenazi S, May-Zahav M, Sulkes J, Zilberberg R, Samra Z. Hygiene behaviour in rural Nicaragua in relation to diarrhea. Int J
Increasing antimicrobial resistance of Shigella isolates in Israel Epidemiol 1998;27:1090—100.
during the period 1984 to 1992. Antimicrob Agents Chemother
1995;39:819—23. 87. Myo Han A, Moe K. Household faecal contamination and diarrhea
risk. J Trop Med Hyg 1990;93:333—6.
73. Guerrant RL, Van Gilder T, Steiner TS, Thielman NM, Slutsker L,
Tauxe RV, et al. Practice guidelines for the management of 88. Mertens TE, Jaffar S, Fernando MA, Cousens SN, Feachem RG.
infectious diarrhea. Clin Infect Dis 2001;32:331—51. Excreta disposal behaviour and latrine ownership in relation to
the risk of childhood diarrhea in Sri Lanka. Int J Epidemiol

74. Wright CE, el Alamy M, DuPont HL, Holguin AH, Hsi BP, Thacker 1992;21:1157—64.
SB, et al. The role of home environment in infant diarrhea in
rural Egypt. Am J Epidemiol 1991;134:887—94. 89. Traore E, Cousens S, Curtis V, Mertens T, Tall F, Traore A, et al.
Child defecation behaviour, stool disposal practices, and child-
hood diarrhea in Burkina Faso: results from a case-control study.
J Epidemiol Community Health 1994;48:270—5.